Objectives: To assess whether clinical input during calls to the NHS 111 telephone-based advice service is associated with lower rates of subsequent emergency department attendance and hospital admission.Design: Although NHS 111 largely employs non-clinical call handling staff to triage calls using computerised clinical decision support software, some support is available from clinical supervisors, and additionally some calls are referred to out-of-hours General Practitioners (GP). We used linked data sets to examine GP and secondary care activity following calls to NHS 111, adjusting for the patient characteristics, signs and symptoms recorded during the NHS 111 call.Setting: Out-of-hours care in three areas of North West London that have an integrated approach to delivering NHS 111 and out-of-hours GP care.Participants: NHS 111 calls for children and young people aged 15 years or under. We excluded calls that were diverted to the emergency ('999') service or where patients were advised to go to an emergency department. This left callers who were either referred to a GP or advised to manage their health needs at home. Primary and secondary outcome measures:The percentage of callers attending any emergency departments, major emergency department, or minor injury unit within ten hours of the NHS 111 call, and the percentage admitted to hospital following visits to emergency departments. Results:Of the 10,356 callers, 2,898 (28.0%) were advised by NHS 111 to manage their health needs at home, with an appointment with an out-of-hours GP made for the remaining 7,458 (72.0%). 14.9% (432/2,898) of the callers who were advised by NHS 111 to manage their health needs at home attended an emergency department with ten hours, compared with 16% (1,207/7,458) of callers who had an outof-hours appointment with an out-of-hours GP. After adjusting for patient characteristics, GP out-ofhours appointment was associated with lower rates of emergency department attendance (adjusted odds ratio, 0.86, 95% CI, 0.75-0.99),). When we subset emergency department types, a GP out-of-hours appointment was associated with lower rates of minor injury unit attendance (adjusted odds ratio, 0.32, 95% CI, 0.23 -0.44) but not major emergency department attendance (adjusted odds ratio 1.06, 95% CI 0.90-1.24). There was no association with hospital admission. Review by an NHS 111 clinical supervisor was associated with fewer emergency department attendances (adjusted OR 0.77, 95% CI, 0.62-0.97).

AimsNHS 111 is a telephone-based advice service that employs nonclinical call-handling staff to triage calls using computerised clinical decision support software. The safety of the service is contested, particularly for children and younger people, and there have been concerns that its introduction has increased emergency department utilisation. We examined predictors of emergency department attendance among out-of-hours NHS 111 callers. MethodsWe studied out-of-hours calls made to NHS 111 for people aged under 15 years in three areas of north-west London between July 2013 and February 2015 (n=11,279), using linked NHS 111, GP and secondary care data. We assessed agreement between the advice by NHS 111 given regarding emergency care and subsequent attendances using Cohen's kappa.Logistic regression tested the association between emergency department attendances and the advice given, patient characteristics, and features of the telephone call and wider healthcare system. We also examined predictors of emergency inpatient admissions. ResultsThe vast majority (87.3%) of NHS 111 calls happened outside of hours. NHS 111 advised 18.5% of callers to attend emergency departments, and 63.8% of these attended within 10 hours. Overall concordance between the advice regarding emergency care and subsequent attendances was moderate (kappa 0.51). Two-thirds of callers were transferred by NHS 111 to out-ofhours general practice care. Callers who spoke to a GP as part of the NHS 111 call episode were less likely to attend emergency departments than other callers (adjusted odds ratio 0.54, 95% confidence interval (CI) 0.47-0.62) but were no less likely to be admitted (adjusted odds ratio 0.95, 95% CI 0.65-1.38). ConclusionsOur finding that patients are around half as likely to attend emergency departments if they speak with an out-of-hours GP requires further scrutiny to address confounding. But, if valid, this study suggests that other areas of England might reduce emergency department utilisation by integrating NHS 111 more closely with out-of-hours GP services. ■

Concerns have been raised that some referrals from NHS 111 to Emergency Departments (EDs) for children are unnecessary; increasing pressure on services whilst some appropriate cases are not being referred. This paper’s aim is to analyse what factors were associated with ED attendances after parents/carers call NHS 111.We studied 32,398 NHS 111 calls made between April 2013 and February 2015 for children aged under 16 in the area covered by three Clinical Commissioning Groups.We used logistic regression on linked NHS 111, EDs and GP Out-Of-Hours (GPOOH) data to test predictors of ED attendance within five hours following NHS 111 contact. The data included patient characteristics (e.g. age, symptom), call characteristics (e.g. time of call) and wider contextual characteristics.Of the patients studied, 13.1% were advised to go to ED[1] and 9.3% visited ED. There was a mismatch between advice and actual attendance: 8.1% of all patients studied were advised to attend ED but did not and 4.3% were not advised to attend but did so anyway. The symptom that patients presented was an important factor in this (see Figures 1 and 2).Abstract G208 Figure 1Patients who attended ED regardless of advice and patients who attended ED after being advised to indexed by total number of patients advised to attend EDAbstract G208 Figure 2The percentage of patients advise to attend ED against the percentages of patients who attended ED per symptom with area of circles representing the number of patients displaying that symptomWe found, using the logistic regression results, distance from an A&E department strongly affected whether patients attended ED; patients were roughly half as likely to attend for every kilometre further away. GP patient satisfaction rating and office hours, also affected the patient’s likelihood of attending ED (see Table 1).Abstract G208 Table 1Adjusted odds ratios from logistic regression analysisAdvice given was more risk averse in recommending ED attendance than the behaviour of patients with injuries, limb problems and burns. In contrast, call handlers did not seem to assuage the parents’ concern regarding other symptoms such as fever. A limitation of this study is that we could not assess the proportion of children whose symptoms changed after the call and that appropriateness of the ED attendance or lack of attendance cannot be assessed.[1] 3.7% were advised by GPOOH and 9.3% were advised by NHS 111. Patients are frequently redirected to GPOOH service by NHS 111 whilst on the line. Patients make little distinction between the services.

Study Objectives: Current guidelines recommend oral anticoagulation (OAC) to reduce stroke risk in high-risk atrial fibrillation (AF) patients. However, emergency department (ED) prescribing is inconsistent. The provider factors influencing OAC prescribing in the ED are unknown. This study aimed to identify factors that prevent and support OAC prescribing for AF by ED physicians.Methods: These results are part of a larger study to identify barriers to optimal atrial fibrillation management in the ED at a tertiary care academic hospital. We completed semi-structured interviews with 18 providers who had treated a patient for new-onset AF in the ED within the previous 30 days. A qualitative, grounded theory framework was applied to data collection and analysis to develop a theoretical foundation to examine prescribing practices among physicians in a single ED setting. Data collection, transcription, and analysis were conducted simultaneously, using lineby-line coding and constant comparative analysis in Atlas.ti, and reviewed for agreement by two analysts. We stratified data by years of experience: 0-4, 5-10 and greater than 10 years post-residency.Results: Two themes emerged in our analysis: (1) OAC prescribing at the time of discharge and (2) variable use of guidelines for OAC prescribing in our analysis. OAC prescribing: Stratification by years of experience demonstrated that providers with less than 10 years of experience are least likely to prescribe anticoagulation at discharge unless the patient was already on an OAC medication. Alternatively, these providers will admit a patient with new onset AF who needs anticoagulation rather than send them home with OAC. Additionally, if they do prescribe, they consult with cardiology first. In contrast, those with greater than 10 years experience frequently discharge patients on OAC, with only a few relying on a cardiology consult. Regardless of experience, ED providers consider multiple issues in the decision-making process for OAC prescribing including the patient's ability to follow-up with their primary care provider, type of insurance, and several social factors. Guideline use: Irrespective of how long they have been in practice, the majority of providers indicated that they are unaware of, and do not typically use, particular guidelines or protocols when treating patients with AF in the ED. Providers that do utilize existing guidelines cited CHADS, CHADS 2 , and CHA 2 DS 2 -VASc, literature by the American Heart Association and American College of Cardiology, Up-to-Date, and various peer-reviewed publications. The major barriers preventing the use of prescribing guidelines were: 1) knowledge of existing guidelines, 2) lack of guidelines appropriate for ED or acute management, and 3) desire to treat patients using individualized medicine.Conclusions: The decision to prescribe OACs is complex and multifocal and involves the transition of care of patients with a chronic condition from the episodic ED care to the outpatient setting. Improving guideline adherence to OA...

No abstract